Predicting major adverse cardiac events in diabetes and chronic kidney disease: a machine learning study from the Silesia Diabetes-Heart Project DOI Creative Commons
Hanna Kwiendacz,

Bi Huang,

Yang Chen

и другие.

Cardiovascular Diabetology, Год журнала: 2025, Номер 24(1)

Опубликована: Фев. 15, 2025

Язык: Английский

Multimodal machine learning in precision health: A scoping review DOI Creative Commons
Adrienne Kline, Hanyin Wang, Yikuan Li

и другие.

npj Digital Medicine, Год журнала: 2022, Номер 5(1)

Опубликована: Ноя. 7, 2022

Abstract Machine learning is frequently being leveraged to tackle problems in the health sector including utilization for clinical decision-support. Its use has historically been focused on single modal data. Attempts improve prediction and mimic multimodal nature of expert decision-making met biomedical field machine by fusing disparate This review was conducted summarize current studies this identify topics ripe future research. We accordance with PRISMA extension Scoping Reviews characterize multi-modal data fusion health. Search strings were established used databases: PubMed, Google Scholar, IEEEXplore from 2011 2021. A final set 128 articles included analysis. The most common areas utilizing methods neurology oncology. Early merging strategy. Notably, there an improvement predictive performance when using fusion. Lacking papers clear deployment strategies, FDA-approval, analysis how approaches diverse sub-populations may biases healthcare disparities. These findings provide a summary as applied diagnosis/prognosis problems. Few compared outputs approach unimodal prediction. However, those that did achieved average increase 6.4% accuracy. Multi-modal learning, while more robust its estimations over methods, drawbacks scalability time-consuming information concatenation.

Язык: Английский

Процитировано

231

Machine learning and deep learning predictive models for type 2 diabetes: a systematic review DOI Creative Commons
Luis Fregoso-Aparicio, Julieta Noguez, Luis Montesinos

и другие.

Diabetology & Metabolic Syndrome, Год журнала: 2021, Номер 13(1)

Опубликована: Дек. 1, 2021

Abstract Diabetes Mellitus is a severe, chronic disease that occurs when blood glucose levels rise above certain limits. Over the last years, machine and deep learning techniques have been used to predict diabetes its complications. However, researchers developers still face two main challenges building type 2 predictive models. First, there considerable heterogeneity in previous studies regarding used, making it challenging identify optimal one. Second, lack of transparency about features models, which reduces their interpretability. This systematic review aimed at providing answers challenges. The followed PRISMA methodology primarily, enriched with one proposed by Keele Durham Universities. Ninety were included, model, complementary techniques, dataset, performance parameters reported extracted. Eighteen different types models compared, tree-based algorithms showing top performances. Deep Neural Networks proved suboptimal, despite ability deal big dirty data. Balancing data feature selection helpful increase model’s efficiency. Models trained on tidy datasets achieved almost perfect

Язык: Английский

Процитировано

115

Predicting Coronary Heart Disease Using an Improved LightGBM Model: Performance Analysis and Comparison DOI Creative Commons
Huazhong Yang, Zhongju Chen,

Yang Huajian

и другие.

IEEE Access, Год журнала: 2023, Номер 11, С. 23366 - 23380

Опубликована: Янв. 1, 2023

Coronary heart disease (CHD) is a dangerous condition that cannot be completely cured. Accurate detection of early coronary artery can assist physicians in treating patients. In this study, prediction model called HY_OptGBM was proposed for predicting CHD by using the optimized LightGBM classifier. To optimize classifier, hyperparameters were adjusted. addition, its loss function improved, and trained adjusted hyperparameters. applying most advanced hyperparameter optimization framework (OPTUNA). The improved referred to as focal (FL). evaluated data from Framingham Heart Institute. evaluate performance model, various metrics, including precision, recall, F score, accuracy, MCC, sensitivity, specificity, AUC, used. AUC value 97.9%, which better than other comparative models. results demonstrate rate identification among general population utilizing method. This, turn, could serve mitigate costs associated with medical treatment patients suffering CHD.

Язык: Английский

Процитировано

69

Machine learning-based models for the prediction of breast cancer recurrence risk DOI Creative Commons
Duo Zuo,

Lexin Yang,

Yu Jin

и другие.

BMC Medical Informatics and Decision Making, Год журнала: 2023, Номер 23(1)

Опубликована: Ноя. 29, 2023

Abstract Breast cancer is the most common malignancy diagnosed in women worldwide. The prevalence and incidence of breast increasing every year; therefore, early diagnosis along with suitable relapse detection an important strategy for prognosis improvement. This study aimed to compare different machine algorithms select best model predicting recurrence. prediction was developed by using eleven learning (ML) algorithms, including logistic regression (LR), random forest (RF), support vector classification (SVC), extreme gradient boosting (XGBoost), decision tree (GBDT), tree, multilayer perceptron (MLP), linear discriminant analysis (LDA), adaptive (AdaBoost), Gaussian naive Bayes (GaussianNB), light (LightGBM), predict area under curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative (NPV) F1 score were used evaluate performance prognostic model. Based on performance, optimal ML selected, feature importance ranked Shapley Additive Explanation (SHAP) values. Compared other 10 results showed that AdaBoost algorithm had successfully recurrence adopted establishment Moreover, CA125, CEA, Fbg, tumor diameter found be features our dataset More importantly, first use SHAP method improve interpretability clinicians based algorithm. offers a clinical identifies cancer.

Язык: Английский

Процитировано

48

Credit Card Fraud Detection for Contemporary Financial Management Using XGBoost-Driven Machine Learning and Data Augmentation Techniques DOI Creative Commons
Teuku Rizky Noviandy, Ghalieb Mutig Idroes, Aga Maulana

и другие.

Indatu Journal of Management and Accounting, Год журнала: 2023, Номер 1(1), С. 29 - 35

Опубликована: Сен. 12, 2023

The rise of digital transactions and electronic payment systems in modern financial management has brought convenience but also the challenge credit card fraud. Traditional fraud detection methods are struggling to cope with complexities contemporary strategies. This study explores potential machine learning, specifically XGBoost (eXtreme Gradient Boosting) algorithm, combined data augmentation techniques, enhance detection. research demonstrates effectiveness these techniques addressing imbalanced datasets improving accuracy. showcases a balanced approach precision recall by leveraging historical transaction employing like Synthetic Minority Over-sampling Technique-Edited Nearest Neighbors (SMOTE-ENN). implications findings for profound, offering bolster integrity, allocate resources effectively, strengthen customer trust face evolving tactics.

Язык: Английский

Процитировано

46

A systematic review and meta-analysis of artificial neural network, machine learning, deep learning, and ensemble learning approaches in field of geotechnical engineering DOI Creative Commons
Elaheh Yaghoubi, Elnaz Yaghoubi, Ahmed A. Khamees

и другие.

Neural Computing and Applications, Год журнала: 2024, Номер 36(21), С. 12655 - 12699

Опубликована: Май 13, 2024

Abstract Artificial neural networks (ANN), machine learning (ML), deep (DL), and ensemble (EL) are four outstanding approaches that enable algorithms to extract information from data make predictions or decisions autonomously without the need for direct instructions. ANN, ML, DL, EL models have found extensive application in predicting geotechnical geoenvironmental parameters. This research aims provide a comprehensive assessment of applications addressing forecasting within field related engineering, including soil mechanics, foundation rock environmental geotechnics, transportation geotechnics. Previous studies not collectively examined all algorithms—ANN, EL—and explored their advantages disadvantages engineering. categorize address this gap existing literature systematically. An dataset relevant was gathered Web Science subjected an analysis based on approach, primary focus objectives, year publication, geographical distribution, results. Additionally, study included co-occurrence keyword covered techniques, systematic reviews, review articles data, sourced Scopus database through Elsevier Journal, were then visualized using VOS Viewer further examination. The results demonstrated ANN is widely utilized despite proven potential methods engineering due real-world laboratory civil engineers often encounter. However, when it comes behavior scenarios, techniques outperform three other methods. discussed here assist understanding benefits geo area. enables practitioners select most suitable creating certainty resilient ecosystem.

Язык: Английский

Процитировано

23

Interpretable Machine Learning Approach to Predict Hepatitis C Virus NS5B Inhibitor Activity Using Voting-Based LightGBM and SHAP DOI Creative Commons
Teuku Rizky Noviandy, Aga Maulana, Irvanizam Irvanizam

и другие.

Intelligent Systems with Applications, Год журнала: 2025, Номер 25, С. 200481 - 200481

Опубликована: Янв. 15, 2025

Язык: Английский

Процитировано

7

A Novel Approach for Polycystic Ovary Syndrome Prediction Using Machine Learning in Bioinformatics DOI Creative Commons
Shazia Nasim, Mubarak Almutairi, Kashif Munir

и другие.

IEEE Access, Год журнала: 2022, Номер 10, С. 97610 - 97624

Опубликована: Янв. 1, 2022

Polycystic ovary syndrome (PCOS) is a critical disorder in women during their reproduction phase. The PCOS commonly caused by excess male hormone and androgen levels. follicles are the collections of fluid developed ovaries may fail to release eggs regularly. results miscarriage, infertility issues, complications pregnancy. According recent report, diagnosed 31.3% from Asia. Studies show that 69% 70% did not avail detecting cure for PCOS. A research study needed save identifying early. main aim our predict using advanced machine learning techniques. dataset based on clinical physical parameters utilized building models. novel feature selection approach proposed optimized chi-squared (CS-PCOS) mechanism. ten hyper-parametrized models applied comparison. Using CS-PCOS approach, gaussian naive bayes (GNB) outperformed state-of-the-art studies. GNB achieved 100% accuracy, precision, recall, f1-scores with minimal time computations 0.002 seconds. k-fold cross-validation accuracy score. model accurate prediction. Our reveals features prolactin (PRL), blood pressure systolic, diastolic, thyroid stimulating (TSH), relative risk (RR-breaths), pregnancy prominent factors having high involvement helps medical community overcome miscarriage rate provide through early detection

Язык: Английский

Процитировано

48

Effective Feature Engineering Technique for Heart Disease Prediction With Machine Learning DOI Creative Commons

Azam Mehmood Qadri,

Ali Raza, Kashif Munir

и другие.

IEEE Access, Год журнала: 2023, Номер 11, С. 56214 - 56224

Опубликована: Янв. 1, 2023

Heart failure is a chronic disease affecting millions worldwide. An efficient machine learning-based technique needed to predict heart health status early and take necessary actions overcome this worldwide issue. While medication the primary treatment, exercise increasingly recognized as an effective adjunct therapy in managing failure. In study, we developed approach enhance detection based on patient parameter data involving learning. Our study helps improve at its stages save patients' lives. We employed nine algorithms for comparison proposed novel Principal Component Failure (PCHF) feature engineering select most prominent features performance. optimized PCHF mechanism by creating new set innovation achieve highest accuracy scores. The newly created dataset eight best-fit features. conducted extensive experiments assess efficiency of several algorithms. decision tree method outperformed applied learning models other state-of-the-art studies, achieving high score 100%, which admirable. All methods were successfully validated using cross-validation technique. research has significant scientific contributions medical community.

Язык: Английский

Процитировано

40

Integrating Genetic Algorithm and LightGBM for QSAR Modeling of Acetylcholinesterase Inhibitors in Alzheimer's Disease Drug Discovery DOI Creative Commons
Teuku Rizky Noviandy, Aga Maulana, Ghazi Mauer Idroes

и другие.

Malacca Pharmaceutics, Год журнала: 2023, Номер 1(2), С. 48 - 54

Опубликована: Июль 20, 2023

This study explores the use of Quantitative Structure-Activity Relationship (QSAR) studies using genetic algorithm (GA) and LightGBM to search for acetylcholinesterase (AChE) inhibitors Alzheimer's disease. The uses a dataset 6,157 AChE their IC50 values. A model is trained evaluated classification performance. results show that achieved high performance on training testing set, with an accuracy 92.49% 82.47%, respectively. demonstrates potential GA in drug discovery process findings contribute by providing insights about allow more efficient screening compounds accelerate identification promising candidates development therapeutic use.

Язык: Английский

Процитировано

36